The Impact of Physician non-English Language Proficiency Level on Cancer Screening Project Summary/Abstract Communication between patients and clinicians is an essential component of cancer screening. Ineffective communication can lead to patient misunderstanding of the importance of cancer screening or disagreement with it, leading to poor adherence. Patients with limited English proficiency (LEP) are at risk of disparities in cancer screening because of impaired communication with their clinicians. True language concordance occurs when clinicians are fluent in the languages their patients speak and offers an important way to reduce disparities for LEP patients. All too often though, there is only partial language concordance between patients and clinicians because the clinician does not speak the patient's language fluently but does not use an interpreter. When there is only partial language concordance, there are concerns for the quality of care provided. The goal upon completion of the proposed studies is to have evidence that will help establish standards for testing of, and allowing practice with, clinician non-English language proficiency and to reduce language-related disparities in cancer screening. The specific aims for this application are (1) To determine if a detailed self-reporting scale of clinician language proficiency correlates to the results of a validated oral proficiency test, (2) To explore clinicia characteristics that are predictors of accuracy in self-assessment of non-English language proficiency; (3) To assess whether breast, cervical and colorectal cancer screening completion rates are higher for LEP patients with truly language concordant clinicians compared to those with partially language concordant clinicians. To achieve these aims, the proposed studies will first compare clinicians' self-assessment of their non-English language proficiency using the Interagency Language Roundtable (ILR) scale compared to the Clinician Cultural and Linguistic Assessment (CCLA), a gold standard, validated oral proficiency test. The two measures will be analyzed using quantile-quantile plots of the two distributions, the ILR and the CCLA. The analysis will also employ Spearman's rank correlation coefficient to assess the consistency of the ILR and the CCLA. Additional analyses will be conducted to assess the relationship between obtaining a passing score on the CCLA and spoken language history and provider demographic variables. Next, exploratory univariate and subsequent multivariate analyses will be performed to ascertain which factors are associated with the over- or underestimation of non-English language proficiency by clinicians. Finally, analyses will be conducted to assess the impact of clinician non-English language proficiency on breast, cervical, and colorectal cancer screening. Hierarchical/multilevel modeling will be employed to account for patient, clinician and system level factors, with separate models conducted for each type of cancer screening. Level one will assess the probability of completing cancer screening at the patient level, including the covariates listed above. Level 2 will account for clinician factors listed above and level 3 will include clinician work sites.